Cognitive-Motivational Behavior Therapy:Retaining Gamblers in Treatment

Edelgard Wulfert, Ph.D.University at Albany – SUNY

e.wulfert@albany.edu

When gambling becomes a problem

Continuum of gamblingNone Occasional Frequent Problem Pathological

l____________l__________l____________l

NRC Classification (1999):

Level 0: Never gambledLevel 1: Social or recreational gamblingLevel 2: At-risk or problem gamblingLevel 3: Pathological gambling (PG)

Pathological gambling (PG)

A psychological disorder characterized by

• a persistent and recurring failure to resist gambling behavior that isharmful to the individual and/or others

• high levels of psychiatric comorbidity

• significant similarities with addictive disorders

Prevalence Rates

Current best estimates: (point prevalence)

Problem gamblers: 3-5% Pathological gamblers: 1.5%

PG is a significant public health problem Treatment development is essential

Treatment of PGNon-completers &

Drop-outsEcheburua et al. (1996)

64 slot machine gamblers (BT, CT, or CBT) 45%

McConaghy et al. (1991)120 mixed gamblers (BT, Relax., Aversion) 47%

Treatment of PG

Non-completers &Drop-outs

Sylvain et al. (1997)29 video poker players (CBT*) vs. WL) 36% *)

Petry et al. (2006)231 PGs (GA, GA+CBT, GA+Workbook)

(Of 8 CBT sessions attended: 7%=0; 32% ≤ 5) 39%(Chapters completed: 30%=0, 34% ≤ 5) 64%

Treatment of PG

• Most studies have shown good treatment effects for gamblers who are retained

• But all studies have also shown significant dropout rates.

This seems to indicate that researchers may pay insufficient attention to motivational factors

Caveats when implementing CBT

Tacit assumption of CBT: Treatment-seeking clients are ready to change

• Addictions are functional (adaptive value) • Ambivalence is a core feature of addiction

• Lack of commitment• Dropout• Relapse

Key to change:

Tipping the motivational balance

Development of CMBT(Cognitive-Motivational Behavior Therapy)

Cognitive-Motivational BehaviorTherapy

CMBT integrates:• motivational enhancement techniques • psycho-education • cognitive & behavior therapy strategies

Goal:• First engage patients in treatment• Then provide insight and skills to foster

behavior change

Treatment Development of CMBT: Phase 1

3 Sessions of Motivationally Enhanced Therapy (modeled after Project Match)

• Personalized feedback from Intake Assessment• Use of MI principles (EE, DD, SS, RR) • Decisional Balance Exercises • Values clarification• Goal setting

CMBT: Phase 2

12-15 Sessions of:

CT (modeled after Ladouceur)• Identifying and correcting distorted beliefs

about gambling and chance events

Psychoeducation • Facts about gambling; odds

Behavioral strategies• Problem solving & skills training • Evaluation of lifestyle and choices

CMBT: Phase 3

2 Sessions of Relapse Prevention(modeled after Ladouceur / Marlatt)

• Stop, look, and listen• Emergency Procedures

Conjoint session with SIGO (where indicated)

Treatment Pilot Study(Wulfert, Blanchard, Freidenberg, Martell, 2005)

22 treatment-seeking male PGs

• Assigned to CMBT (9) or TAU (12)• Mean age 43 (29-59)• Avg. length of gambling 15 yrs (3-30)• Mean DSM criteria 8 (7-10)• Mean SOGS score 16 (9-20)

Main Outcomes

• Validity Check of Motivational Intervention• Assessed after Session 3• Significant increase in clients’ motivation

and readiness to change

• Main Outcomes• DSM-IV Characteristics• SOGS Scores

Pre/Post Treatment Gambling Severity

DSM-IVSOGS

15.9

1.2

14

7.8

0

17

Pre Post

8.1

1.3

7.5

4.8

0

10

Pre Post

Exptl.Control

[F(1,15)17.61, p=.001] RM Anova TimeXCond [F(1,15) 14.1, p = .002]

Treatment Retention

CMBT TAURetained in Tx: 9/9 (100%) 8/12 (67%) *

* X2 = 8.05, p = .005

Patients in CMBT:• Completed treatment and 12-month follow-up• Maintained treatment gains in follow-up• Showed decreases in depression and state anxiety• Showed heart rate decreases to gambling stimuli

DSM-IV and SOGS Scores: CMBT

02

46

810

1214

1618

20

Pre Post 3 mos. 6 mos. 12 mos.0

2

4

6

8

10

Pre Post 3 mos. 6 mos. 12 mos.

DSM-IV CriteriaSOGS Scores

* RMA: Time: F(4,5) 29.96, p =.001

HR (BPM) Pre - Post Treatment

2.031.69

0.130.31

Gambling Scene 1 Gambling Scene 2

BL

corr

ecte

d th

e B

MP

Pre PrePost Post

**

* p<.05

(Freidenberg, Blanchard, Wulfert, Malta, 2002)

Limitations

• Small sample size • Non-randomized control group• No follow-up data on control group• No process measures

Controlled follow-up study isneeded

NIMH-funded Treatment Development Study

RCT with 46 treatment-seeking PGs Randomly assigned to • CMBT (n=23; 16 men, 7 women) • GA (n=23; 16 men, 7 women)

Demographic Information

• Age: mean 44 years (range 24 - 70)• Ethnicity:

85% Caucasian• Education:

76% at least high school or some college• Marital status:

57% married; 24% single; 19% sep/div./wid. • Employment:

76% fulltime; 9% unemployed• Household income:

Median: $35 - 50K (Range: <$10K to >$100K)• Gambling debt:

Median: $10K (Range: $500 - $65K)

CMBT: 12 Session Manualized Tx

• 3 Sessions of Motivational Enhancement• 8 Sessions of CBT• 1 Session of Relapse Prevention

A motivational interviewing style is employed throughout treatment

3 master’s level therapists (CSWs)

Gamblers Anonymous Control Group

• Clients referred to GA were instructed to attend weekly GA meetings

• Patient advocate

Main Outcomes & Assessments

Main Outcome variables• DSM criteria, SOGS, Money lost gambling, Days

gambled

Secondary Outcome variables• Readiness to change; cognitive distortions

Assessments• Pre / Post / 3-month / 6-month follow-up• CMBT process variables: also at 4 and 8 weeks

AttritionCMBT:• 1/23 (4.3%) dropped out after Session 2• 22/23 (95.7%) attended all 12 sessions• 1/23 (4.3%) was lost to 6-month follow-up GA:• 10/23 (43.5%) never attended any meetings• 14/23 (60.9%) attended <3 meetings • 8/23 (34.8%) were lost to follow-up assessmts.

Fisher’s exact test (dropouts): p<.001

Preliminary Outcomes

• GA was similarly effective to CMBT for gamblers who attended GA meetings regularly• Problem: High rate of noncompliance and

dropout and from GA

• Intent-to-treat analyses• Last assessment point carried forward

0

2

4

6

8

10

12

14

Pre Post 3 mos 6 mosSO

GS

Scor

es (0

-20)

GA

CMBT

DSM-IV Criteria and SOGS Scores

DSM-IV Diagnosis of PG SOGS

0

20

40

60

80

100

Pre Post 3 mos 6 mos

Perc

ent m

eetin

g PG

dia

gnos

is

GA

CMBT

* Group Diff’s: p <.01

0

20

40

60

80

100

Pre Post 3 mos 6 mos

Dol

lars

Gam

bled

(% P

re)

0

20

40

60

80

100

Pre Post 3 mos 6 mosD

ays

Gam

bled

(% P

re)

GA

CMBT

GA

CMBT

Dollar Amount and Number of Days Gambled(percent from baseline)

Money lost gambling Days gambled

Group Diff’s: p <.01

CMBT Process Measures

• Readiness to Change (URICA)• Session 4 Scores correlated with

treatment outcome

• Irrational Cognitions (GBQ)• Session 8 Scores correlated with

treatment outcome

Conclusions

MBCT• Retains patients in treatment• Increases motivation to change• Decreases irrational beliefs re. gambling• Decreases gambling behavior• Possibly decreases urges and arousal

Limitations & Future Directions• Promising, but empirical support is modest

at this time• 1 pilot study + 1 RCT = 32 CMBT patients

• Positive effects are limited to 1 single setting• Test of transportability is necessary

• High dropout rate from GA• Test against a more stringent control group is

necessary

• Plan: • Conduct a large2-site RCT with stringent controls

Acknowledgements:

Co-investigator: SUNY AlbanyDr. Edward Blanchard

Former students: Current students:Dr. Julie Hartley Ms. Christine FrancoDr. Marlene Lee Ms. Ruthlyn Sodano

Ms. Kristin HarrisMs. Bianca Jardin

Collaborator:Dr. Carlos Blanco, NYPI

Therapists and Patients Center for Problem Gambling, Albany, NY